NID

Natural Interception Dataset

Purpose

This repository stores the natural interception dataset (NID). The system used to intercept the electro-magnetic leakage emitted by a screen is detailed in the ICASSP20 paper named Electro-Magnetic Side-Channel Attack Through Learned Denoising and Classification" --> ArXiv or IEEE Xplore

Repository Content

• data
  • train: 424 images based on BSD432 samples:
    • in: eavesdropped samples
    • ref: reference grayscale samples
  • val: 68 BSD68 samples
    • in: eavesdropped samples
    • ref: reference grayscale samples
  • missing: the 8 ref samples missing in the train dataset
    • ref

Licensing

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

The dataset is free to use until you propagate the licence and you cite our paper using the following format:

@article{lemarchand2019electro, title={Electro-Magnetic Side-Channel Attack Through Learned Denoising and Classification}, author={Lemarchand, Florian and Marlin, Cyril and Montreuil, Florent and Nogues, Erwan and Pelcat, Maxime}, journal={arXiv preprint arXiv:1910.07201}, year={2019} }

Contact

Florian Lemarchand : [email protected], A=florian.lemarchand and B = insa-rennes